Measuring software-defined radios
Rick Nelson, Chief Editor -- Test & Measurement World, 11/1/2006
Software-defined radios (SDRs) are combining the flexibility of digital signal processing (DSP) with RF circuitry to allow software to dynamically control communications parameters such as carrier frequency, bandwidth, power levels, and data rate. In addition, the software at the heart of these “digital RF” systems can perform functions such as filtering signals, establishing modulation and coding schemes, and determining frequency-hopping patterns.
 |
|
A digital RF radio employs software running on a DSP to dynamically control communications parameters. |
The emergence of SDRs is driven by techniques such as the direct up-conversion to RF from the output of a digital-to-analog converter (DAC). Such techniques allow more of the analog hardware to be integrated, but they reduce signal access, burying within silicon the test points to which you could once attach your oscilloscope or spectrum analyzer to make traditional measurements such as occupied bandwidth, channel power, adjacent channel power, error vector magnitude (EVM) and correlated power, and rise and fall time.
Furthermore, SDRs require tests beyond these traditional RF transceiver conformance tests and the like to include software-regression testing to ensure that software-controlled operations don't inject unwanted glitches, interference, pulse aberrations, and software-dependent phase errors.
In addition, SDRs often employ nonlinear digitally enhanced power amplifiers that rely on software to perform crest-factor reduction, implement digital predistortion, and provide digital feedback linearization, further complicating measurement tasks.
To provide guidance for SDR troubleshooting, Tektronix has just released an application note that gives tips for applying logic analyzers, oscilloscopes, and real-time spectrum analyzers to the task. The app note provides details on frequency-hopping measurements and the use of a frequency-mask triggering function to identify glitches.
It also describes how to locate crosstalk from microprocessors operating near the 2.4-GHz ISM band, and it provides hints on identifying transients caused by adaptive modulation and coding schemes, such as the scheme employed in HSDPA to switch on-the-fly between QPSK and 16-QAM.
Click to download the 16-page note, “Software Defined Radio Testing Using Real-Time Spectrum Analysis.”
